05:17 <GitHub24> [jruby] jakago opened issue #4689: __dir__ does not return canonicalized absolute path https://git.io/vQZLQ

05:55 <kares_> what's the status of considering a drop for Java 7 with 9.2 ?

05:56 <kares_> (have seen 9.1 twice slower on 7 than 8 on a Rails app ... probably due indy and I am not sure its worth the effort to 'fix')

08:47 <travis-ci> kares/jruby (test-set-native:09a504d by kares): The build passed. (https://travis-ci.org/kares/jruby/builds/247393098)

08:49 <GitHub174> [jruby] kares created test-set-native (+17 new commits): https://git.io/vQZW8

08:49 <GitHub174> jruby/test-set-native df27212 kares: get native SortedSet working (using Java's TreeSet as an order backend)

08:49 <GitHub174> jruby/test-set-native 7ecc71f kares: extract out Array's Comparator impls into classes (for reusability)

08:49 <GitHub174> jruby/test-set-native 2ce8f3f kares: a well working (first) draft of rewritten native Set impl (based on set.rb)

09:58 <GitHub91> [jruby] kares pushed 1 new commit to master: https://git.io/vQZzx

09:58 <GitHub91> jruby/master d3a5ee2 kares: revert previous fix for detecting alias-ed method's defined class...

10:34 <GitHub164> [jruby] kares opened pull request #4690: native implementation of set.rb (master...test-set-native) https://git.io/vQZwl

10:41 <chrisseaton> kares: I would have thought that if people had managed to upgrade to 7 they would have managed to upgrade to 8, and people who are 'stuck behind' are generally so on 6, but I'm guessing

10:42 <GitHub130> [jruby] kares closed issue #2089: Invalid encoding with nokogiri https://git.io/vQZrI

10:45 <travis-ci> jruby/jruby (master:d3a5ee2 by kares): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/247451779)

10:48 <GitHub43> [jruby] kares closed issue #1593: Seeing undefined method `_validation_callbacks' with Rails 4.0.4 https://git.io/vQZr2

11:18 <GitHub52> [jruby] patch0 opened issue #4691: `syslog/logger` from the Ruby standard library is missing in JRuby 9 https://git.io/vQZiz

12:27 <travis-ci> jruby/jruby (test-set-native:09a504d by kares): The build passed. (https://travis-ci.org/jruby/jruby/builds/247428621)

15:58 <codefinger> I'm getting "warning: ambiguous Java methods found, using subscribe(rx.functions.Action1)". But Action is a FunctionalInterface, and the arg I'm passing in is a block, like `thing.subscribe { |x| do_stuff }`. Is there a way to disambiguate it without mucking up my Ruby code too much?

17:21 <headius> kares_: it may be indy, or other slower aspects of Java than 7...we have not chosen to drop 7 yet though

17:21 <headius> we recommend 8 but we still have people running 7 for apps out there

17:52 <headius> eregon: how does TR handle this LJE case?

17:53 <headius> my naive fix searching stack for the target scope was too eager

19:21 <GitHub32> [jruby] headius created fix_break_lje (+1 new commit): https://git.io/vQnSL

19:21 <GitHub32> jruby/fix_break_lje 203daad Charles Oliver Nutter: Mark blocks as escaped and raise LJE for break when appropriate....

19:30 <GitHub194> [jruby] headius pushed 1 new commit to fix_break_lje: https://git.io/vQn9b

19:30 <GitHub194> jruby/fix_break_lje ccd8534 Charles Oliver Nutter: Additional fix for escaped blocks, set in optimized interp path....

20:54 <GitHub71> [jruby] headius pushed 1 new commit to ruby-2.4: https://git.io/vQnjd

20:54 <GitHub71> jruby/ruby-2.4 c2a7ea2 Charles Oliver Nutter: [2.4] Fixes for Numeric recursion guard logic in ec800e6....

21:34 <GitHub26> [jruby] headius pushed 1 new commit to fix_break_lje: https://git.io/vQckT

21:34 <GitHub26> jruby/fix_break_lje 9d5dcee Charles Oliver Nutter: Only set literal closures as escaped after their call returns.

22:03 <GitHub64> [jruby] enebo pushed 1 new commit to master: https://git.io/vQcqv

22:03 <GitHub64> jruby/master f2e7b82 Thomas E. Enebo: Store token list of strptime format string in a cache so we do not continually...

22:20 <chrisseaton> headius: what's LJE?

22:21 <headius> LocalJumpError